https://www.irjet.net/archives/V9/i4/IRJET-V9I4455.pdf

1. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 04 | Apr 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 3276
Voice Controlled News Web Based Application With Speech
Recognition Using Alan A.I.
Aditya Sharma1, Subiya Siddiqui2, Sugandha Trivedi3, Ali Murtaza4, Mohd Faiez5, Shyam
Dwivedi6
1, 2,3,4,5 Student of B. Tech fourth year, Department of Computer Science and Engineering, Rameshwaram Institute of
Technology & Management, Lucknow, India
6Assistant Professor & Head of Department, Department of Computer Science and Engineering, Rameshwaram
Institute of Technology & Management, Lucknow, India
A decade ago, we used to consume news via newspaper,
media and by communicating with each other but with
recent advancement in artificial intelligence, natural
language processing and computing power we can
access news only via voice command which will make
our news gathering experience too much fun and
interactive. With rapid development in computer
technology, we have reached much closer to
advancement in artificial intelligence and natural
language processing. Natural language processing helps
to analyse the audio and return the needed response in
electronic voice. Our software sends the source audio to
Alan AI cloud server where voices are analysed so that
response can be returned. Our web app based
application makes news gathering fun and interactive
and voice assistant makes it so easy to listen to the news.
News presents facts of society and world and to know
about the current situation, one must look for news that
comes in many forms like newspaper, electronic news,
media, but none of them provide comfort to the listener,
but our ALAN AI provides comfort by retrieving the
news which is commanded by the customer through
voice.
2. PROJECT OBJECTIVES
Traditional news gathering methods were quite complex
and were time taking and also very costly. That’s why we
have built a project which provides news in more
interactive and fun way. User just has to invoke the
assistant by specific key words after wards our web app
will provide news through voice assistant. So, anyone
who cannot get news by reading the news articles or any
traditional method, our project is very useful and
advanced method for getting news.
3. RECENT DEVELOPMENT OF AI
Recently, humanity has witnessed technological
advancement in the artificial intelligence field.
Previously we used news and media channels to interact
with news but now with advancement in computing
power in mobile phones we can interact with news all
across the globe through mobile phones. But the whole
level of experience of interacting with news can be
upgraded by utilising the artificial intelligence and
subfields of artificial intelligence like natural language
processing, robotics, machine learning, expert system,
neural network.
4. BRANCHES OF AI:
4.1. MACHINE LEARNING
In [1] Machine Learning we study computer algorithms.
The algorithm improves itself by experience and data.
Machine learning algorithms are trained by large
amounts of data to do a task without being explicitly
trained to do. Complex problems are solved
economically using machine learning algorithms.
----------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - To move into the future we have to keep
updating ourselves and keep adapting to the latest
technologies. If we observe the updating in news industry
we would notice that we still read news through paper
which is bad for environment but we have many digital
tools like Artificial Intelligence, Natural Language
Processing, Speech Recognition System The proposed idea
of speech recognition system which provides news only
with the help of user command through voice with very
detailed or concise manner according to user’s choice. The
user can access news by category, popular news channel,
by terms etc. The web app is completely responsive in
mobile phone, tablet, and laptop and desktop. The project
is developing using reactJS, JAVASCRIPT, Visual Studio
Code, and Alan AI.
Key Words - React.js, Node.js, Mongo-DB, Visual Studio
Code, ALAN AI.
1. INTRODUCTION

2. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 04 | Apr 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 3277
Fig. 1- Simple ML Pipeline
4.2. NEURAL NETWORK
[2]Neural network is a way to mimic the function of the
human brain. In the human brain there are connections
of billions of neurons which are connected with each
other by synapses and neurotransmitters which help to
store memory and help to retrieve the information at the
moment of need.
In artificial neural networks there are node layers in
which each neuron is connected by each other and there
are many hidden layers and also contain one output
layer. Each node has some threshold value and
associated weight. If data in node exceeds threshold
value then data is transferred to another node.
Fig. 2 – Neural Network
4.3. ROBOTICS
Robotics3 learning is the intersection of machine
learning and robotics, robots are built to do a specific
task which is very complex for humans like lifting towers
or manufacturing cars and with the help of machine
learning algorithms robots can improve their efficiency
and precision to do a task. Robotics is an emerging field
and many developments happening in robotics which
improve precision, are also very economical to use.
Today every industry is using robotics to manufacture
their products.
Fig. 3 - Robotics
4.4. EXPERT SYSTEM
Expert [4] system is the way to emulate decision making
similar to the human expert. Expert system is code to
solve complex problems through the body of if-then-else
rule through conventional problems. Expert system was
first artificial intelligence software which worked
successfully. The expert system has two parts. The
inference engine and knowledge base. The knowledge
base represents facts and rules. The inference engine
completed its task by gathering the facts to deduce
upgraded facts. Inference engine includes debugging
abilities and explanation.
Fig. 4 – Expert System
4.5. NATURAL LANGUAGE PROCESSING
Natural Language Processing 5 is a branch of artificial
intelligence which helps to process the language and
respond to the query of the user in voice. NPL
summarises the large volume of texts rapidly, even in
real time. You have also interacted with NLP through
Alexa, Google assistant, speech to text dictation and
customer service chat bots.
Fig. 5 - NLP

3. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 04 | Apr 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 3278
5. TOOLS USED
5.1. ALAN AI
Alan A.I. 6 is the advanced A.I based voice platform that
allows you to add voice assistance to your web app
without overhead.
Fig. 6 – ALAN AI
5.2. REACT.JS
React (also known as React.js or ReactJS)[7] is a free and
open-source front end library for building user
interfaces based on UI components.
5.3. NODE.JS
Node.js is an open-source JavaScript runtime
environment and library to run web applications outside
the client's browser. Node.js uses asynchronous, event
driven model and is perfectly used for data-interactive
app. Node.js is used in making real time web application
network application, also in distributed system.
5.4. JAVA SCRIPT
JavaScript is a high-level, compiled language that
conforms to the ECMA Script standard which confirms
the interoperability in web pages. It has dynamic typing,
prototype-based object orientation, and first class
function. It is multi-paradigm, supporting event driven,
functional, and imperative programming style. JavaScript
possesses application programming interfaces (APIs) for
working with text, dates, regular expression, standard
data structure, and the Document Object Model (DOM).
5.5. VISUAL CODE STUDIO
Visual Studio Code is a source-code editor that can be
used with a variety of programming languages, including
Java, JavaScript, Go, Node.js, Python and C++. It is based
on the Electron framework, which is used to develop
Node.JS applications that run on the Blink Layout
Engine.
Some features are:
● Intelligent code completion
● Code refactoring
● Support for debugging
● Syntax Highlighting
● Embedded Git
● Snippet
Fig. 7 – Visual Studio Code
6. IMPLEMENTATION
Project implementation can be divided into three
parts:
● Frontend development
● API connection with frontend
● ALAN AI backend programming
6.1. FRONT END DEVELOPMENT
Frontend Development is a part which is done by using
react.JS and material UI.React.JS is used because of large
data set and change in data set does not require
reloading the page. The rendering process is done by
Virtual Dom which is very fast. The purpose of render ()
function is to display the specified HTML code inside the
specified HTML element. A render function read props
and state and return our JSX code to the root component
of our app. The virtual DOM is programming concept
where an ideal or virtual representation of UI is kept in
memory and synced with real DOM by a library such as
react DOM. The fast rendering required in the project
like this where the user is interacting with web app
using actual voice. Material UI on other hand is a popular
react ui frame work. It allows developers for quick
development by providing various components.

4. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 04 | Apr 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 3279
Fig. 8 – User Interface
6.2. API CONNECTION WITH FRONTEND
The Application Programming Interface, API is a
messenger that takes requests and tells the system what
user wants and then return the response hence it is
software that provides a foundation for connection
between two applications. This connection is used to
send a request and return a response to the user. An API
is the real backend connectivity engine between various
other applications.
Two APIs are used in this project. Each web app has its
API which connects with another application. The two
APIs were from news API and ALAN AI’s API. One can
search and find news across the globe using news APIs. A
lot of query search are available, you can retrieve news
by country, accident, and pandemic or on games
according to developer choice.
6.3. ALAN AI BACKEND PROGRAMMING
Alan Studio provides great support to developers by
offering free voice assistants. Alan offers SDKs for
different platforms to embed a voice assistant to your
app. Voice Scripts are written in JAVASCRIPT which
make them highly customizable and flexible.
Alan AI provides a voice assistant to any type of
application only with the help of APIs. ALAN AI does all
the backend work of voice assistant developers only
have to connect the api with its web app to import the
features of ALAN AI.
6.3.1. ALAN STUDIO
Alan AI is a powerful web-app based integrated
development environment where you can debug, write
and test dialog scenarios for your voice Chabot or
assistance.
Fig. 9 – ALAN Voice Assistant
6.3.2. ALAN CLIENT SDKs
Alan’s SDK provides an efficient way to embed a feature
in any existing application.
6.3.3. ALAN CLOUD
The ALAN AI backend framework consists of
Speech Synthesis; Natural Language Processing and
Automatic Speech Recognition. The Alan cloud contains
all sufficient data to perform the voice assistant task in a
very flexible manner.
7. CONCLUSION
Traditionally reading newspaper used to take a lot of
time and it was not interactive and fun to read
newspaper with the help of our project.
Gathering news will become more fun and interactive.
The project is capable of retrieving news of all genres
across the globe, and it will retrieve news according to
user command if the user wants to listen to news in
concise form or in depth.
8. FUTURE DEVELOPMENT
To move ahead further we can add more functionality in
our application. As soon as ALAN AI will update their
software with more languages like Hindi and Japanese,
then our web can also take command in many different
languages and can be used globally. We also planned to
implement recommendation system using machine
learning which will recommend news according to user’s
interest and we will also classify the news according to
category of news, like if user will give command for
positive news, then our web app provide news which
will lie in positive category and vice-versa.

5. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 04 | Apr 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 3280
9. REFERENCE
[1] Machine learning | Wikipedia
https://en.wikipedia.org/wiki/Machine_learning
[2] Neural Network | Ibm.com|
https://www.ibm.com/cloud/learn/neural-
networks
[3] Robotics | Wikipedia | en.wikipedia.org
[4] Expert System | Ibm.com |
https://www.ibm.com/cloud/learn/IBM
[5] Natural Language Processing | IBM |
www.ibm.com
[6] ALAN AI | https://alan.app
[7] React | JavaScript library | https://reactjs.org
[8] Node.js | https://nodejs.org
Mohd Faiez - He is currently a student of
B. Tech fourth year, Dept. of Computer
Science and Engineering, Rameshwaram
Institute of Technology & Management,
Lucknow and working on News Reading
App by using ALAN A.I.
Ali Murtaza - He is currently a student of
B. Tech fourth year, Dept. of Computer
Science and Engineering, Rameshwaram
Institute of Technology & Management,
Lucknow and working on News Reading
App by using ALAN A.I.
Aditya Sharma - He is currently a
student of B. Tech fourth year, Dept. of
Computer Science and Engineering,
Rameshwaram Institute of Technology &
Management, Lucknow and working on
News Reading App by using ALAN A.I.
Sugandha Trivedi - She is currently a
student of B. Tech fourth year, Dept. of
Computer Science and Engineering,
Rameshwaram Institute of Technology &
Management, Lucknow and working on
News Reading App by using ALAN A.I.
Subiya Siddiqui – She is currently a
student of B. Tech fourth year, Dept. of
Computer Science and Engineering,
Rameshwaram Institute of Technology &
Management, Lucknow and working on
News Reading App by using ALAN A.I.
Shyam Dwivedi – He is currently
working as an Assistant Professor and
Head of Department in Rameshwaram
Institute of Technology & Management,
Lucknow, India. He is M. Tech – 2012
BIT Mesra, Ranchi, he has a teaching
experience of 10 years and 1-year in
TCS Industrial experience.
10. BIOGRAPHIES